PaperHub

官方获奖论文

PaperHub 收录的每个会议、最新一届的官方杰出论文 / 亚军奖，仅限主会议赛道。

NeurIPS
NeurIPS 20256 篇
8.7
21

Gated Attention for Large Language Models: Non-linearity, Sparsity, and Attention-Sink-Free

杰出论文NeurIPS 2025Zihan Qiu et al.
8.2
20

1000 Layer Networks for Self-Supervised RL: Scaling Depth Can Enable New Goal-Reaching Capabilities

杰出论文NeurIPS 2025Kevin Wang et al.
8.2
20

Why Diffusion Models Don’t Memorize: The Role of Implicit Dynamical Regularization in Training

杰出论文NeurIPS 2025Tony Bonnaire et al.
10.0
23

Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model?

亚军NeurIPS 2025Yang Yue et al.
9.1
16

Optimal Mistake Bounds for Transductive Online Learning

亚军NeurIPS 2025Zachary Chase et al.
8.2
22

Superposition Yields Robust Neural Scaling

亚军NeurIPS 2025Yizhou Liu et al.
ICLR
ICLR 20263 篇
8.0
12

LLMs Get Lost In Multi-Turn Conversation

杰出论文ICLR 2026Philippe Laban et al.
7.0
13

Transformers are Inherently Succinct

杰出论文ICLR 2026Pascal Bergsträßer et al.
8.0
18

The Polar Express: Optimal Matrix Sign Methods and their Application to the Muon Algorithm

荣誉提名ICLR 2026Noah Amsel et al.
ICML
ICML 20256 篇
8.5
9

Roll the dice & look before you leap: Going beyond the creative limits of next-token prediction

杰出论文ICML 2025Vaishnavh Nagarajan et al.
7.8
11

Conformal Prediction as Bayesian Quadrature

杰出论文ICML 2025Jake C. Snell et al.
7.8
11

Train for the Worst, Plan for the Best: Understanding Token Ordering in Masked Diffusions

杰出论文ICML 2025Jaeyeon Kim et al.
7.8
13

CollabLLM: From Passive Responders to Active Collaborators

杰出论文ICML 2025Shirley Wu et al.
7.2
14

Score Matching with Missing Data

杰出论文ICML 2025Josh Givens et al.
7.2
9

The Value of Prediction in Identifying the Worst-Off

杰出论文ICML 2025Unai Fischer-Abaigar et al.
COLM
COLM 20254 篇
7.8
14

Fast Controlled Generation from Language Models with Adaptive Weighted Rejection Sampling

杰出论文COLM 2025Ben Lipkin et al.
7.3
18

Shared Global and Local Geometry of Language Model Embeddings

杰出论文COLM 2025Andrew Lee et al.
7.3
15

Don’t lie to your friends: Learning what you know from collaborative self-play

杰出论文COLM 2025Jacob Eisenstein et al.
6.8
16

Hidden in plain sight: VLMs overlook their visual representations

杰出论文COLM 2025Stephanie Fu et al.